The renewal quote came in 18% higher: what to do before the term ends

The timeline decides the outcome

A renewal quote 18% above last year is an opening position, not a decision, and the calendar decides who can move it. The date that matters is not the term end date. It is the notice deadline in the auto-renewal clause, because once that passes the commercial question is closed. A buyer who starts at 30 days is negotiating inside the window the vendor counted on.

The mechanics are published, and they favour the prepared buyer. Microsoft's billing documentation says recurring billing is on by default for card and billing-profile subscriptions, so a subscription keeps being billed at the end of each term unless someone turns it off, and that some subscriptions can only be cancelled inside a limited window after purchase or renewal. The same pages describe the exit path in days: expired, disabled, deleted, with data held for a set period and then removed.

The calendar below is the working version. The 120-to-60-day row is where the money is; the later rows exist to stop a slipped date becoming an unplanned multi-year commitment.

The data to assemble before the call

Renewal conversations are lost on vagueness, not on price. The vendor knows the seat count it billed; the buyer needs the count in use, by department and by module, beside the contracted quantity. A one-page report of licensed seats, active users, the gap, and the modules nobody launched turns a percentage demand into a scope question.

The unused-seat list is the strongest document in the room because it is arithmetic. It is also why subscription rationalisation has moved up the agenda: the inventory of what is owned, used and paid for is the substance of the negotiation.

Add the true-up history and the contract terms: what last year's true-up cost, when it was triggered, and which line defines the uplift or the overage rate. Vendor licensing pages are the place to start, and Atlassian's documentation shows how plainly the subscription model can be set out, but the answer that binds is in your own order form. The buying route matters too: Microsoft treats direct, enterprise agreement, volume licensing and cloud solution provider purchases differently, and grace periods vary by program.

The clauses that matter at signature

A renewal can only be as good as the contract it renews. Six clauses decide what happens when a quote arrives above last year's number, and each is either useful or decorative.

The auto-renewal notice period is the first. Thirty days gives a buyer a month to run a process; ninety gives a quarter. The second is the annual uplift, which should be a percentage cap rather than an uncapped reference to a public index. The third is price protection across a multi-year term: without an explicit hold, a multi-year commitment carries an annual increase that resets the value of the term.

The last three surface only when something goes wrong. Overage and true-up pricing determines what extra usage costs, and pricing that is not in the agreement is pricing the vendor sets later. Termination for convenience, even a narrow version, preserves the option to stop paying for something the organisation stops using. Data at exit is the last: who exports what, in which format, in what window, at what cost. Vendor pages answer part of that. After cancellation, customer data may be deleted after a set period and no later than a stated outer limit, and buying the same subscription again does not restore it.

The levers that work, and the ones that do not

The levers that move a number are trades. A multi-year term for a capped annual uplift is the strongest, because it gives the vendor the booking it wants and the buyer predictability. Re-tiering seats to the count in use, and dropping modules nobody signed into, shrinks the base the uplift applies to. Annual prepay buys a discount on a service worth keeping. A documented competitive alternative, meaning a real quote, changes how a deal is priced.

Two structural levers are worth pricing even when they look awkward. A reseller can open a discount path the vendor's direct motion does not, at the cost of a layer between buyer and vendor. A committed-spend agreement trades a spending floor for better unit rates: it suits a buyer who can forecast consumption honestly.

The approaches that fail are consistent. Threatening to leave with nowhere to go is read as a bluff once the vendor checks whether an alternative is in flight. Bluffing about a competing quote collapses in the second call. Negotiating only with the account executive stalls when the discount sits with the deal desk, so ask who approves pricing exceptions. Revealing the budget before the scope is fixed hands the vendor a ceiling.

Getting internal agreement before the call

The vendor call is the easy part. The hard part is arriving with one position. The business owner has to state which capabilities are load-bearing and which are tolerated, in writing, because a renewal that cuts a module someone depends on creates a second problem. Finance has to agree the target number, the maximum acceptable increase, and the no-deal plan.

Give the group one owner for the negotiation and a stated approval path for the final number. Two people sending different signals to the same account team is the fastest way to lose the discount on offer.

Calendar discipline for the next renewal

A renewal that surprised anyone is a process failure that repeats annually unless something changes. The fix is unglamorous: one register holding every contract's term end, notice deadline, uplift language, owner and expected value, with a task raised 120 days before each deadline. Notice dates come from the signed order, not memory, because the two disagree more often than anyone expects.

Keep the evidence current so the next negotiation starts with data rather than a scramble. A quarterly review of active users against licensed seats, and of module adoption against the modules paid for, keeps the unused-seat list a standing document. Watch pricing direction too, not only the invoice: published list pricing and plan changes move before a renewal quote arrives, and the vendor's own pricing page is a cheap way to check whether an uplift reflects a change in the product or simply a new number.